What is Google Shopping

The name covers three separate things that are worth keeping apart. The first is Google's Shopping tab, a storefront where shoppers compare items from several retailers. The second is Shopping ads, the paid format auctioned inside Google Ads and shown on Search, on Google Images and on the tab itself. The third is free product listings, which occupy non-advertising slots and carry no cost per click.

All three share the same origin: the product information a retailer deposits in Merchant Center. That platform stopped being called Merchant Center Next in July 2026 and went back to the short name, with no change to accounts, product data or campaigns.

Unlike a text ad, nobody here writes a headline or picks keywords. The system assembles the listing from the image, the title, the price and the store name already present in the submitted data. A fault in the feed is therefore fixed at the data source, and rewriting the campaign does not repair it.

The circuit starts in Merchant Center, where the retailer registers a data source holding its products. Google validates every item against the product data specification and against its policies. An item that passes validation becomes eligible for free product listings; if the Merchant Center account is also linked to a Google Ads account and a campaign includes the item, it enters the paid auction as well.

When somebody searches, Google compares the query with the item's attributes. Title, description, category and product type do the work that keywords would do in a search campaign. Hence the practical rule: relevance is edited in the feed.

Two campaign types coexist on that foundation. Standard Shopping campaigns split the catalogue into listing groups and allow bids by brand, category, id or custom labels. Performance Max takes the same catalogue and extends it to YouTube, Gmail, Display, Discover and Maps with automated bidding. If a product sits in two Standard Shopping campaigns for the same country, campaign priority decides which one bids; when the higher priority campaign runs out of budget, the next one bids.

With no positive keywords, the available control looks different: negative keywords to exclude queries, listing groups for segmentation, inventory filter, custom labels and, above all, the quality of the data you send.

The first decision that depends on this is where to spend the effort. In a search campaign, better results usually mean touching keywords, bids and copy. In Shopping, the same improvement almost always runs through the catalogue: a title that opens with brand and model, a clean image, a correct category, a product identifier that is actually there.

The second is about allocation. Free product listings appear on Search, on the Shopping tab, in Google Images, on YouTube, in Maps, in Lens and in Gemini at no cost per click. A retailer who only watches the Google Ads account leaves that traffic out of the analysis and attributes it wrongly. Before raising budget it pays to measure how much visibility already exists for free.

The third is about risk. Merchant Center disapproves items and can suspend the account when the submitted data does not match the store. A suspension switches off ads and free listings at the same time, so feed maintenance decides whether the whole channel keeps running.

• Write title in the order a buyer uses: brand, model, distinguishing attribute and, where relevant, size or colour. The attribute allows up to 150 characters, so put the distinctive part first.
• Send gtin whenever the manufacturer assigns one. Google recommends it explicitly, and mpn only takes its place when the product has no GTIN.
• Split the catalogue with custom labels by margin, turnover or season before you build the campaign structure. Useful segmentation is defined in the data.
• Review the Merchant Center diagnostics weekly and clear warnings before the deadline expires, because an unattended warning ends in a disapproval.
• Measure free product listings separately from ads. The Merchant Center performance report distinguishes them, and mixing the two credits the bid with results it did not earn.
• Use campaign priority only when two Standard Shopping campaigns genuinely compete for the same product in the same country.

• Treating Shopping like a search campaign and expecting positive keywords to govern exposure. In Shopping only the negative ones exist.
• Leaving the brand out of the title on the grounds that it already travels in brand. The query is compared with the item's text attributes, and the title is the most visible of them.
• Assuming an approved item will be shown. For the paid side it must also sit in an active campaign with budget available.
• Duplicating the catalogue across several campaigns without setting priorities, which leaves the effective bid at the mercy of budget allocation.
• Switching off Performance Max during a sale to regain control, without first measuring which share of performance came from surfaces Standard Shopping does not cover.